A detailed description of the sequence of deformation steps leading to phase inversion during compounding in a low-viscosity-ratio co-polyester/polyethylene blend is presented. Visualization using a glass window and sampling of the blend at different mixing times enabled identification of the intermediate morphologies of the major component en route to phase inversion. Based on these observations, a theoretical model is developed to predict the time to phase inversion. The model incorporates a simplified flow-field approximation and the calculation of strain imparted to the major component domains. A strain-based criterion for phase inversion is then proposed, which, in conjunction with the model, yields an explicit expression for the time to phase inversion during compounding, t(P.I). The model predictions are seen to be in good agreement with the increase of t(P.I.) on scaleup between two mixing bowls. The correct functional dependence of t(P.I.) on the nominal maximum-shear-rate is predicted. Using combination of pure drag and planar extensional flow, the model predictions are shown to be consistent with the observed dependence of t(P.I.) on the volume fraction of the minor component and the blend viscosity ratio.